In 2016, the Center for Medicare & Medicaid Innovation initiated the Home Health Value-Based Purchasing (HHVBP) Model in nine states to test the impact of providing financial incentives to home health agencies (HHAs) for improvements in quality of care. A new report found modest improvements in certain quality measures, modest declines in some but not all aspects of utilization, and an overall reduction in Medicare spending.
The analysis, conducted by the Arbor Research Collaborative for Health and L& M Policy Research, focuses on the experience of home health patients and agencies through 2018, the third performance year of the HHVBP Model, and the first year that agencies in the HHVBP states received a payment adjustment.
Overall, the findings for the first year of HHVBP payment adjustments (2018) were similar to those for the two earlier years of the model (2016-2017). There was no consistent evidence across quality, utilization, and spending measures of successively larger impacts of HHVBP due to the application of the payment adjustments in 2018.
The model involves Arizona, Florida, Iowa, Massachusetts, Maryland, Nebraska, North Carolina, Tennessee, and Washington. The HHVBP Model Medicare payments to all eligible agencies in the nine selected states are adjusted upward or downward using a budget-neutral method based on their Total Performance Score (TPS), a composite score of an agency’s quality achievement/improvement.
The amount of the Medicare payment adjustment for each agency is determined by comparing its TPS score with scores for other agencies in the same state. The adjustment process redistributes Medicare payments among agencies within a state to reward agencies with relatively higher achieved quality or improved quality and reduce payments to agencies with lower levels of performance.
Here are some key findings of the evaluation from their report:
HHA Total Performance Scores are higher in each of the first three years of the model. TPS scores serve as broad indicators of HHA performance and are the basis for adjusting Medicare FFS payments to agencies in the nine model states. For each of the first three years of the model (2016-2018), TPS scores for agencies in HHVBP states were higher overall relative to the TPS scores calculated for agencies in the non-model states.
Evidence of reductions in unplanned hospitalizations and use of skilled nursing facilities, but an increase in emergency department use. Through the first three years of HHVBP, evaluators found a modest impact of the model on the claims-based utilization measures that apply to fee-for-service beneficiaries receiving home health services. This includes declines of 0.21-0.30 percentage points in unplanned hospitalization rates among first and all home health episodes, which corresponds to a 1.3 percent to 1.8 percent decrease from average measure values pre-HHVBP implementation.
Declines in overall Medicare spending for FFS beneficiaries receiving home health services appear largely to be explained by reduced spending for inpatient and SNF services. Through the first three years of the model, researchers detected a 1.2 percent decline in average Medicare expenditures per day among FFS beneficiaries in HHVBP relative to the comparison group during and within 30 days following home health episodes. This overall decline can be explained by the observed slower rate of growth in HHVBP states relative to the non-HHVBP states in spending during home health episodes (rather than in the subsequent 30 days). The average annual reduction in total Medicare spending during and within 30 days following home health episodes for FFS beneficiaries receiving home health care in the model is $141 million.
Modest gains in quality of care include greater improvements in functional outcomes. There is a strong pattern through the first three years of the model of relatively small positive effects of HHVBP on many of outcome measures used to calculate TPS scores. This includes a measure of discharge to the community and several measures of improvement in functional status.
Interestingly, agencies interviewed in both HHVBP and non-HHVBP states reported that working with accountable care organizations (ACOs) and managed care plans presented challenges to their operations and care delivery, including restrictions on the number of visits and time-consuming authorization and appeals processes. Agencies reported that these factors negatively impacted agency performance scores and often resulted in reduced patient satisfaction. Despite these challenges, agencies continue to work with managed care plans and ACOs to maintain good relationships with referrers and stay competitive in their respective markets.
In its conclusion, the report states that “through the end of the first year in which the TPS scores, as measures of overall quality performance, were used to adjust Medicare payments to home health agencies, there is no clear evidence that these payment adjustments led to a more pronounced impact of the model on quality of care, utilization, or Medicare spending. Increases in the magnitude of adjustments in future years may change this finding.”